Increases in local blood flow accompany almost all increases in neuronal activity in the healthy brain. These dynamic changes in local blood flow are essential for normal brain function, while also providing the contrast detected in functional magnetic resonance imaging (fMRI). Understanding neurovascular coupling is therefore important both for interpretation of fMRI data, and to understand how the brain regulates its energy supplies in health and disease. While a number of different cellular models of neurovascular coupling have been proposed, our work during the first funded period of this R01 project revealed a new component involved in neurovascular coupling in the brain: the vascular endothelium. Our experiments demonstrated that the vascular endothelium serves to propagate stimulus-evoked dilatory signals along blood vessels in the brain, making it an essential part of healthy neurovascular coupling. Based on our findings to date, we propose that involvement of the vascular endothelium could explain anomalies in earlier astrocyte and / or pericyte-based models of neurovascular coupling, as well as spatiotemporal non-linearities observed in the fMRI BOLD response. In addition, since endothelial dysfunction is involved in a range of systemic pathologies, from cardiovascular disease to diabetes, our findings suggest that these same conditions could directly affect dynamic neurovascular coupling in the brain, providing a possible direct link between systemic microvascular dysfunction and neurodegeneration. In this renewal application we propose to build upon our recent findings, and further explore the role of endothelial signaling in neurovascular coupling. In aim 1 we will investigate how endothelial signaling is initiated in the intact, in-vivo brain. By isolating the location and properties of initiation of endothelial signaling we will explore novel pathways by which neuronal activity could drive vasodilation, while also defining the role of astrocytes and pericytes in our new model. In aim 2 we will determine the types of endothelial signaling pathways involved in the propagation of vasodilation in the brain. These studies will define the sensitivities of neurovascular coupling to endothelial dysfunction caused by disease or pharmacological agents. Mathematical modeling will be used to assimilate mechanistic findings and validate whether they predict the spatiotemporal non-linearities of the hemodynamic response, and thus the fMRI BOLD signal. Finally, we will perform studies to assess the impact of impaired endothelial function on neuronal function, behavior and long-term neurodegeneration. While a causal link between impaired neurovascular coupling and neurodegeneration is to be expected, these effects have never been quantified either acutely or longitudinally, and are of heightened significance if systemic endothelial dysfunction is a risk factor. All experiments will involve the co-development of innovative new in-vivo optical imaging, microscopy and photo-manipulation techniques along with mathematical modeling, image analysis and novel transgenic models of neurovascular control and dysfunction.